Aditi Raghunathan
Unlabeled Data Improves Adversarial Robustness
Yair Carmon, Aditi Raghunathan, Ludwig Schmidt, John C. Duchi, Percy S. Liang
We demonstrate, theoretically and empirically, that adversarial robustness can significantly benefit from semisupervised learning. Theoretically, we revisit the simple Gaussian model of Schmidt et al. [41] that shows a sample complexity gap between standard and robust classification. We prove that unlabeled data bridges this gap: a simple semisupervised learning procedure (self-training) achieves high robust accuracy using the same number of labels required for achieving high standard accuracy.
Unlabeled Data Improves Adversarial Robustness
Yair Carmon, Aditi Raghunathan, Ludwig Schmidt, John C. Duchi, Percy S. Liang
We demonstrate, theoretically and empirically, that adversarial robustness can significantly benefit from semisupervised learning. Theoretically, we revisit the simple Gaussian model of Schmidt et al. [41] that shows a sample complexity gap between standard and robust classification. We prove that unlabeled data bridges this gap: a simple semisupervised learning procedure (self-training) achieves high robust accuracy using the same number of labels required for achieving high standard accuracy.